The DSM-III-R, a major revision of the American Psychiatric Association's diagnostic manual for mental disorders, includes many changes in the Substance Use Disorders section that are designed to enhance the system's usefulness for treatment and research with drug abusers. Because the revised criteria were adopted without extensive prior field testing, little is known about DSM-III-R's actual reliability and validity or about its performance in comparison with other systems for diagnosing substance use disorders. In this study, data will be gathered in a variety of clinical and non-clinical settings with the goals of (a) guiding investigator's and clinicians' use of the criteria and (b) identifying areas in need of revisions in future manuals. Six questions will be addressed: 1. How do DSM-III-R criteria compare with other diagnostic systems (e.g., DSM-III, (ICD-10), quantity/frequency indices, and other problem indicators in diagnosing individuals with and without substance use disorders? 2. How does DSM-III-R compare with other diagnostic systems in terms of interrater agreement, test-retest reliability, predictive validity, and diagnostic stability? 3. How adequate is the construct validity of DSM-III-R criteria as evaluated according to propositions of the Drug Dependence Syndrome construct on which they are based? 4. How valid is the procedure of making DSM-III-R substance use diagnoses on the basis of a single clinical interview? 5. Can we identify clinically useful subgroups of DSM-III-R drug abusers on the basis of dependence severity, severity of coexistent psychopathology and family history of substance use disorders? 6. Can brief screening instruments be found or developed to identify individuals likely to meet DSM-III-R criteria substance use disorders?